INSIGHTS | P E R S P E C T I V E S

REFERENCES

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

R. Lozano et al., Lancet 380, 2095 (2012). C. J. Murray et al., Lancet 380, 2197 (2012). www.who.int/nmh/events/2014/outcome-document.pdf M. Ezzati, E. Riboli, Science 337, 1482 (2012). www.un.org/ga/search/view_doc.asp?symbol=A/66/ 83&Lang=E T. R. Frieden, JAMA 311, 133 (2014). P. Puska, T. Ståhl, Annu. Rev. Public Health 31, 315 (2010). D. A. Chokshi, T. A. Farley, N. Engl. J. Med. 367, 295 (2012). M. Ng et al., Lancet 10.1016/S0140-6736(14)60460-8 (2014). J. S. Lin et al., Ann. Intern. Med. 10.7326/M14-0130 (2014). J. C. de Ruyter et al., N. Engl. J. Med. 367, 1397 (2012). R. Moodie et al., Lancet NCD Action Group, Lancet 381, 670 (2013). S. Basu et al., Am. J. Public Health 103, 2071 (2013). A. Guthrie, “Mexico soda tax dents Coke bottler’s sales,” Wall Street Journal, 26 February 2014. R. Bonita et al., Lancet 381, 575 (2013). 10.1126/science.1259809

1244

SCIENCE COMMUNICATION

Legislators learning to interpret evidence for policy Training for making health policy that has the best impact on a population By Anne Cockcroft,1 Mokgweetsi Masisi,2 Lehana Thabane,3 Neil Andersson4,5

S

hould decision-makers allocate budgets for health services, using the best available evidence? Although the answer seems obvious, most resource allocation is unrelated to evidence of what works best. The well-recognized knowledge-to-action gap (1) has spawned a complex industry of lobbying and knowledge translation. Bridging the gap is a business opportunity for lobbyists, who help to translate evidence in ways that best present a particular interest. Partly to offset these selective, if not frankly biased, interpretations, the last decade has seen an industrial-scale thrust toward systematic reviews to synthesize available evidence (2, 3). But systematic reviews mostly systematize what has been published. For many decisions on health services, there is simply no evidence with the quality one gets from a randomized controlled trial (RCT). Thus, policy decision-makers have to find their way through evidence of varying quality and relevance, only rarely packaged to clarify the population health impact of different choices. A 2012 international forum on evidence-informed health policy-making in low- and middle-income countries called for building the capacity of potential research users to evaluate and use research evidence (4). There are, of course, many levels of research users among health policy-makers and decision-takers. To date, initiatives to build the capacity of research users to interpret evidence have not included actual legislators themselves; instead, they target technical officers and advisers on the assumption that they will then advise the elected representatives appropriately (5). We will describe an effort to help legislators themselves understand the elements of the evidence they need to make decisions (6). AN EXPERIENCE IN BOTSWANA. The Bo-

tswana parliament has 61 members, 57 of them elected and 4 appointed by the head of state. The minimum educational requirement is 7 years of schooling; before the 2014 elections, all members had at least schoolleaving certificates (12 years), and many had

university degrees. Hoping to make parliamentary debates more grounded in evidence, the minister responsible for the national HIV response in Botswana asked CIET Trust, an international research and training group, to train members of parliament (MPs) on the use of research evidence (6). Research assistants contacted MPs by telephone or in person to ask about their needs and interest in training about healthrelated research evidence. They managed to interview roughly half of the MPs; none refused, but some were unavailable within the allotted time frame. Almost all of those interviewed said they needed better access to reliable evidence and wanted training in how to interpret and evaluate evidence. They cited jargon and unexplained technical language and statistical terms as key blocks to their understanding and use of evidence. They complained of not enough research assistants and lack of those who are available.

“After the training, [it was] noted that parliamentary debate … was more sophisticated and focused on evidence.” Training organized under the auspices of the National AIDS Coordinating Agency was then offered to MPs from all parties on two mornings, with a similar session a year later at the request of the MPs. The training focused on three key areas: the value of counterfactual evidence (having a control or comparison group); how biases can distort findings and reports; and the need for evidence about the impact on a population of policies, rather than change in individual risk. For presentation of evidence, the training emphasized the number needed to treat (NNT) to prevent one adverse outcome, and the unit cost per case saved, in preference to parameters of individual impact like relative risk or simply cost of activities. One session explained research terminology; items sciencemag.org SCIENCE

12 SEP TEMBER 2014 • VOL 345 ISSUE 6202

Published by AAAS

Downloaded from on March 27, 2016

looking to low- and middle-income countries as growth markets for their products. For instance, between 1997 and 2009, the annual growth rate of volume consumption per person for soft drinks was 2.4% in high-income countries and 5.2% in lowand middle-income countries (12). A crossnational analysis of 75 countries showed that each 1% rise in soft drink consumption was associated with an additional 4.8 overweight adults per 100 and an additional 2.3 obese adults per 100 (13). Out of concern for the public health effects of these trends, Mexico introduced a tax on soft drinks and “junk” food in 2013. Preliminary evidence indicates that the tax is helping to curb consumption. For example, the Coca-Cola Company reported reduced sales volumes of 5 to 7% attributed to the tax (14). More generally, policy approaches to counter the marketing of unhealthy products include limits on advertising, regulating locations of sale, and counter-advertising. Ultimately, global efforts to prevent NCDs depend on successful national and local implementation. As an example, Brazil, in response to the 2011 U.N. political declaration, launched a national NCD plan of action. The plan includes promoting smoke-free environments, sharply increasing cigarette taxes, and banning tobacco advertising; signing agreements with the food industry to reduce sodium content and eliminate trans fats; and promoting and studying urban physical activity interventions (15). Meeting the global “25 by 25” goal of NCD prevention will require rapid dissemination of proven interventions like MPOWER, as well as investment in designing and evaluating new approaches to promoting physical activity, reducing harmful alcohol use, and altering unhealthy diets. This in turn will require governments to redirect their public health systems toward population-level environmental interventions that prevent NCDs. ■

COULD IT WORK ELSEWHERE? A first

Botswana parliamentarians discussing evidence.

PHOTO: MOSES MAGADZA

legislators asked about most were in a RCT: control or comparison group, statistical significance, number needed to treat, and bias. A hands-on session asked participants to review the value of examples of health research evidence for health policy-making. The scenarios summarized from published articles ranged from small studies with no control group, to large, well-performed RCTs (see the photo). Of 54 elected representatives, 36 attended one or both of the training sessions, including seven ministers, the deputy speaker, the leader of the opposition, and the chair of the parliamentary committee on health and HIV and AIDS. The lively and engaged group reported that concepts like NNT and cost per case saved had immediate resonance for them. Training legislators about evidence will not transform health policy-making overnight. Having access to good evidence and the ability to interpret it are not enough to ensure that policies are evidence-based. Many other considerations come into play; studies have documented hindrances to evidence-informed health policy-making in both low- and high-income countries (7, 8), and systematic reviews identified many barriers to health policy-makers’ using evidence, such as competing influences (9, 10). 1

CIET Trust Botswana, Gaborone, Botswana. 2Minister of Presidential Afairs and Public Administration, Government of Botswana, Gaborone, Botswana. 3Epidemiology and Biostatistics, McMaster University, Hamilton, Canada. 4Family Medicine, McGill University, Montreal, Canada. 5Center for Investigation of Tropical Diseases (CIET), Universided Autónoma de Guerrero, Acapulco, Mexico. E-mail: [email protected]

Training legislators to interpret and evaluate research evidence is uncharted territory. The feedback from the Botswana legislators was very favorable; they asked for further sessions to cover the topics in more detail and for the training to be offered to other decision-makers. After the training, one of the authors (M.M.), a senior parliamentarian, noted that parliamentary debate—for example, around the updated national HIV policy—was more sophisticated and focused on evidence. But it is early yet, and we have not attempted any formal measurement of the longer-term impact of the training. Thus far, we have demonstrated the feasibility and acceptability of this sort of training and have increased our understanding of what elected representatives want to learn and what they can cope with. If elected representatives had a good understanding of the evidence they need to support rational decision-making, they could start to demand different kinds of evidence. For example, they might start to require those who submit budgets, or researchers receiving government funding, to produce evidence about population benefits in addition to individual benefits, about the number needed to treat to prevent one adverse outcome, and about the unit costs per case saved of different programs. They might push to see the results of systematic reviews and research syntheses. Or they might at least demand studies with acceptable counterfactual evidence. They could become active and informed parties in setting the research agenda.

SCIENCE sciencemag.org

step for researchers to make this a reality is to get beyond the idea that legislators have neither the background nor the interest to learn about how to interrogate and use evidence to support decision-making. Several factors contributed to success in Botswana and could be relevant to similar training elsewhere. Seeking the views of the MPs ahead of the training made them partners in the effort and guided the level and content of the sessions. To increase relevance of the learning materials, we used examples from Botswana or the southern Africa region. Linking the training to a particular topic—in this case, it was HIV prevention policies—increased its immediate relevance. Other factors contributed in Botswana and might be less easy to replicate elsewhere. Botswana has a small and functional parliament. In another setting, major conflicts within the legislature could make multiparty training such as this very difficult to implement. We also had a powerful champion in the Minister for Presidential Affairs and Public Administration, who had received master’s training in evidence-based planning, as part of a program to build the capacities of researchers in southern Africa (11). Identifying a similar key supporter could help when attempting to introduce training for MPs in other parliaments. ■ REF ERENCES AND NOTES

1. A. Haines, S. Kuruvilla, M. Borchert, Bull. World Health Organ. 82, 724 (2004). 2. D. Moher, A. Liberati, J. Tetzlaff, D. G. Altman, PLOS Med. 6, e1000097 (2009). 3. J. Lavis et al., J. Health Serv. Res. Policy 10 (suppl. 1), 35 (2005). 4. Report on International Forum on Evidence Informed Health Policy in Low- and Middle-Income Countries, Addis Ababa, Ethiopia, 27 to 31 August 2012; http:// global.evipnet.org/wp-content/uploads/2013/02/ Addisreport2012.pdf. 5. World Health Organization, Evidence-informed policymaking: SURE (2013); www.who.int/evidence/sure/en/. 6. A. Cockcroft, M. Masisi, L. Thabane, N. Andersson, J. Publ. Health Policy 10.1057/jphp.2014.30 (2014). 7. J. Court, J. Young, Evidence Policy 2, 439 (2006). 8. C. J. Jewell, L. A. Bero, Milbank Q. 86, 177 (2008). 9. S. Innvaer, G. Vist, M. Trommald, A. Oxman, J. Health Serv. Res. Policy 7, 239 (2002). 10. L. Orton, F. Lloyd-Williams, D. Taylor-Robinson, M. O’Flaherty, S. Capewell, PLOS ONE 6, e21704 (2011). 11. R. Geneau et al., Retrovirology 6 (suppl. 3), P94 (2009). ACKNOWL EDGMENTS

The training of Botswana legislators was part of the African Development AIDS Prevention Trials capacity (ADAPT) program, a CIET Trust program carried out with support from the Global Health Research Initiative—a research funding partnership composed of the Canadian Institutes of Health Research; Department of Foreign Affairs, Trade and Development Canada (DFATD), and the International Development Research Centre (IDRC); supported by a grant from IDRC, Ottawa, Canada—and with the financial support of the Government of Canada provided through DFATD. 10.1126/science.1256911 12 SEP TEMBER 2014 • VOL 345 ISSUE 6202

Published by AAAS

1245

Legislators learning to interpret evidence for policy Anne Cockcroft et al. Science 345, 1244 (2014); DOI: 10.1126/science.1256911

This copy is for your personal, non-commercial use only.

If you wish to distribute this article to others, you can order high-quality copies for your colleagues, clients, or customers by clicking here.

The following resources related to this article are available online at www.sciencemag.org (this information is current as of March 27, 2016 ): Updated information and services, including high-resolution figures, can be found in the online version of this article at: /content/345/6202/1244.full.html A list of selected additional articles on the Science Web sites related to this article can be found at: /content/345/6202/1244.full.html#related This article cites 9 articles, 2 of which can be accessed free: /content/345/6202/1244.full.html#ref-list-1 This article appears in the following subject collections: Scientific Community /cgi/collection/sci_commun

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright 2014 by the American Association for the Advancement of Science; all rights reserved. The title Science is a registered trademark of AAAS.

Downloaded from on March 27, 2016

Permission to republish or repurpose articles or portions of articles can be obtained by following the guidelines here.

Science Communication. Legislators learning to interpret evidence for policy.

Science Communication. Legislators learning to interpret evidence for policy. - PDF Download Free
482KB Sizes 0 Downloads 6 Views